Additive Manufacturing (AM) has been identified as a key enabler for Mass Customization (MC) due to its negligible tooling cost associated with producing one-off items. This is especially valuable for the medical industry where the ability to create patient-specific products can greatly improve performance and comfort. However, the use of AM so far has only been limited to previously custom-made devices due to the prohibitive design costs associated with a knowledge-intensive and highly manual design process. The research community has often overlooked this area and as yet no study has shown a completely automated process that can reduce or even eliminate this design cost for existing mass-produced ergonomic products (e.g. respirators). This study investigates the methodology of developing a completely automated design pipeline through a case study on Continuous Positive Airway Pressure (CPAP) mask. Through a parametric design approach, a fully automated pipeline was constructed based on a large-scale statistical shape model “learnt” from 9,663 high-resolution facial scans. The pipeline accepts a single “in-the-wild” facial image as the only data input and produces a CAD model of CPAP mask in under a minute. The significant reduction in design time, ease of data acquisition and the complete removal of a manual CAD modelling process can make AM more accessible for CPAP masks manufacturers. The same workflow can potentially be employed to construct automation pipelines for other types of wearables, therefore encouraging the adoption of AM for MC of a wider selection of products.